Over the past six months, Bond Bryan Architects has committed both time and money to investigating in a thorough manner the issues surrounding data exchange. This has been focused on detailed research testing export from our chosen design authoring tool to a variety of solutions. So why did we do this research?
Well for the past 18 months or so we have faced increasing questions around the fact that we don’t use a particular piece of authoring software. No prizes for guessing which one.
At the same time a concerted campaign has been made to highlight the industry need for an open solution through the ‘Open BIM’ campaign. Of course the industry ‘wedge’ diagram that is used to show the progression from 2D to fully collaborative BIM highlights the need for a collaborative format in the shape of IFC (Industry Foundation Classes) at Level 3. Level 3 is not yet fully defined and for many of us the need for IFC is now, at Level 2.
IFC has been given somewhat of a bad name, as i see it over the past two years, and it had become almost common place for people to throw in a slide in their presentations about lost geometry or lost data using IFC as an exchange method. On the flip slide there has been much campaigning on the moral reasons for an open solution. I think even the sceptics get the reasoning behind Open BIM but remain to be convinced it actually works and is a reliable exchange method. The fundamental question therefore for ‘live’ projects is; ‘does it really work?’
Data exchange is ‘perceived’ as poor using IFC and as an industry we need a fully collaborative format to exchange data efficiently between the vast array of software available both now and emerging in the construction industry. The biggest issue has become that our industry is in danger of being split along software divisions when BIM was designed to bring us together for the greater good of our industry. We were meant to be trying to break down silos but we are simply creating different silos. Remember Paul Morrell’s message, “Integrate You Bastards” back in 2011?. Without talking about the real issues and how to solve them we endanger the whole ethos of collaboration and what BIM aims to deliver.
However, we felt as a practice that whilst we believed wholeheartedly in an open solution we weren’t really getting anywhere simply talking about the justification for Open BIM. As i said earlier most people support the concept. We needed cold hard facts and to find out exactly where the issues were. There has been too much ‘finger pointing’ from both sides. These tests would inform our future and determine where we would go next. We felt that if we tackled the issue head on then we would be able to have a sensible discussion about whether IFC works or not.
To give you some context around why we are now looking at these issues, since about 2006 we have sent IFC files regularly to others and never had any issues reported to us. We had also received files from others and again not had anything to worry about. Then the BIM ‘seminar season’ started after the Government’s announcement around BIM back in 2011 and we began to see issues being presented more and more regularly. The question was were these really a case of IFC being as bad as people making out? Was it user error? Was it the exporting software? Was it the receiving software? Was it so bad that we might have to consider alternative solutions?
So towards the end of 2012 we set out to carry out thorough testing of our file export into the software solutions listed earlier. Alongside this we began reporting issues to the software companies and we also had help from others around the world including the Netherlands and Australia (social media has an amazing habit of putting you in touch with the right people). We have now completed a large chunk of our testing and plan on doing more. To date we have tested walls, curtain walls, slabs, columns and spaces. These files also included some windows, doors, openings and niches. This is a fairly decent representation of geometry and data that provides us with an insight of what is going on. We presented much of this work at BIM Show Live 2013 as part of a collaborative presentation with Evolve Consultancy looking at IFC from three different view points.
So what are the conclusions to date? Well without going into the nth amount of detail in this piece, yes there are issues. But it has to be said that having worked hard to understand IFC exchange there were actually far fewer issues than even we had expected when we set off on this journey. Most of the issues we found are with methods we would never need in our day-to-day work and whilst there remain issues that need to be addressed it isn’t going to prevent us exchanging files with others. Most of the outstanding issues are glitches that need resolving by the respective software houses. In the meantime we know about them and we can work round them.
But what we are totally clear about is the existing standard (IFC2x3) if applied correctly by software companies and used correctly by users is suitable for the vast majority of projects. The only exceptions may be extremely complicated geometry which will be improved by the IFC4 standard and probably appear on less than 2% of projects.
The truth is that IFC requires a detailed understanding to get the best out of it. You have to understand how it works and you have to understand both the software you are exporting from and the software you are exporting to. This is part of the reason we employed another member of staff to extend our knowledge of other software. We believe we understand it thoroughly enough now to know that we can create methodologies and approaches that will mean our geometry and data exchange is robust. Part of the reason for wanting to share our knowledge at BIM Show Live 2013 (and subsequently at ThinkBIM in Leeds) is because we want to help inform adult discussion about the subject. Collaboration and the adoption of BIM is hard enough without software getting in the way or for us to spend time arguing about technology issues.
We feel like our research has allowed us to somewhat ‘bridge the divide’ of those who are sceptical about IFC and those who support Open BIM. There is still of course much work to do and many more people to convince that IFC can and does work. We have been asked to present at The Institute of Mechanical Engineers on the subject of interoperability later in the year, and also London Revit User Group (LRUG) which means we can talk more about collaboration and understanding rather than simply being separated down software lines. Most importantly we are seeing others willing to talk and share knowledge and this bodes well for the future.